victor michael zammit
> given a dataset with 100 variables,of five observations in
> each,where the
> 5th observation goes from 1 to 100, e.g. :
>
>
> a1 a2..... a50...... a100
> 1 1 1 1
> 2 2 2 2
> 3 3 3 3
> 4 4 4 4
> 1 2 50 100
> The dataset is saved as A1toA100
> I would like to postfile the standard error,the lower
> bound and the upper
> bound for each variable,at 95% confidence.I am using Stata
> 6.0.What I am
> doing is this case is to program the following ,with the
> ensuing result:
>
> quietly {
> capture program drop pstfle
> program define pstfle
> use A1toA100,clear
> keep A`1'
> quietly ci
> di r(se) " " r(lb) " " r(ub)
> end
> }
> pstfle 1
> pstfle 2
> pstfle 50
> pstfle 100
>
>
> . pstfle 1
> .58309519 .58106823 3.8189318
>
> . pstfle 2
> .50990195 .98428524 3.8157148
>
> . pstfle 50
> 9.5131488 -14.412735 38.412735
>
> . pstfle 100
> 19.506409 -32.158474 76.158474
>
> After highlight "copy" and "paste" ,I use "edit" the data
> in a variables'
> format.
> Is there a more efficient way of doing this ?
I am not clear where you want the results to go.
Although you refer to postfile your code just
displays results in Stata and has no connection
to the -postfile- command. No problem there,
except that the word is a keyword to Stata people.
You could do just this:
(1)
use A1toA100
local i = 1
while `i' <= 100 {
qui ci a`i'
di r(se) " " r(lb) " " r(ub)
local i = `i' + 1
}
I think that should work in Stata 6.
Or you could do this:
(2)
use A1toA100
set obs 100
gen se = .
gen lb = .
gen ub = .
local i = 1
qui while `i' <= 100 {
ci a`i'
replace se = r(se) in `i'
replace lb = r(lb) in `i'
replace ub = r(ub) in `i'
local i = `i' + 1
}
l se lb ub
Note in particular that there is no need
to read the data in again and again.
Nick
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
